|Publication number||US7936694 B2|
|Application number||US 11/729,864|
|Publication date||May 3, 2011|
|Priority date||Apr 3, 2006|
|Also published as||US20070230361|
|Publication number||11729864, 729864, US 7936694 B2, US 7936694B2, US-B2-7936694, US7936694 B2, US7936694B2|
|Inventors||Jonmejoy Das Choudhury|
|Original Assignee||Hewlett-Packard Development Company, L.P.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (11), Referenced by (25), Classifications (33), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application claims priority from Indian patent application 609/CHE/2006, filed on Apr. 3, 2006. The entire content of the aforementioned application is incorporated herein by reference.
With the number of users of Voice over IP (VoIP) systems on the increase, many service providers wish to give a warranty for the service they provide concerning transmission and fault performances to their customers. Furthermore, there is a need to derive a quantitative measure of the quality of the service delivered, to monitor the service and to investigate root causes of any faults occurring in a network. This will enable SLAs (service level agreements) for customers to be shaped, monitored and their violations compensated. The reason for doing this is to mitigate major market inhibitors for most VoIP deployments.
VoIP is a telephony application where the IP network facilitates packet data transport for discrete encoded samples of the voice, as well as signaling information exchanged between entities involved in this service. In order to obtain a good audio quality, it imposes strict real-time and network performance requirements that must be met by the IP network as closely as possible. This is especially important for the sessions set up for transmission of voice.
In the Internet world, however, since services are based on a network service model (=the services realized on the network layer), which is also referred to as “best effort model”, no guarantees are provided concerning quality-of-service of transmission of data streams. The Internet protocol neither provides a bandwidth guarantee nor does it guarantee that the transmission is without any loss. Furthermore, data packets may arrive in any sequence at the destination, there is no guaranteed transmission time and no indication about congestion. In this respect, the Internet network service model differs, for example, from service network models provided in ATM (asynchronous transfer mode) networks. One network service model is, for example, a CBR service model. CBR (constant bite rate) enables a constant transmission rate and a loss-less transmission rate. The sequence of data packets is preserved and congestion does not occur. Further service network models are “variable bit rate (VBR)”, “available bit rate (ABR)” and “unspecified bit rate (UBR)” which provide different quality-of-service which all provide higher quality than IP's “best effort” service model.
The evolution of Internet and ATM network service models also reflects their origins. With the concept of virtual channels (VCs) as a central organization principle, ATM cannot deny its origins from telephony (in which “real circuits” are used). A network based on virtual circuits is arguably more complex than a datagram network. (A datagram is a self-contained packet, one which contains enough information in the header to allow the network to forward it to the destination independently of previous or future datagrams. The terms “datagram” and “packet” are synonymously used herein.) In a datagram network, datagrams only carry their source and destination address and are sent from one router to another, whereby each router knows how to forward the datagrams. Telephone networks, by necessity, had their complexity within the network, since they were connecting dumb end-system devices such as rotary telephones.
The Internet as a datagram network, on the other hand, grew out of the need to connect computers together. Given more sophisticated end-system devices, the Internet architects chose to make the network-layer service model as simple as possible. Additional functionality, for example, in-order delivery, reliable data transfer, congestion control, and DNS name resolution is then implemented at a higher layer, in the end systems.
With regard to VoIP projects, this means that mechanisms for monitoring quality-of-service are established in order to enable a VoIP provider and/or a network provider to give a warranty for the service she/he provides to a client.
A warranty for a service is usually contractually agreed upon by means of an SLA. The TeleManagement Forum's SLA Management Handbook defines an SLA as “[a] formal negotiated agreement between two parties, sometimes called a service level guarantee. Typically, it is a contract (or part of one) that exists between the service provider and the customer, designed to create a common understanding about services, priorities, responsibilities, etc.”
Historically, service level agreements arose in the early 1990s as a way of measuring and managing quality-of-service (QoS) that IT departments and service providers within private (usually corporate) computer networks delivered to their internal customers. To this end, network management tools have been developed, whereby many of the tools rely on the simple network management protocol (SNMP) which is part of the Internet protocol suite and is located on the application layer. SNMP is a client-server based management protocol in which a client queries variables which are stored in local databases, commonly referred to as management information bases (MIB), of the components to be monitored. It should be mentioned that the MIBs are at least partly configured by the manufacturer of the device; some manufacturers do not support MIBs.
A method is provided of monitoring a packet-switched network via which real-time data is transmitted. The method includes sniffing data packets containing real-time data by a monitor subagent to monitor a quality-of-service parameter, notifying, in response to a threshold breach of the quality-of-service parameter, a monitor agent about the breach, notifying a network node manager about the breach, and performing a root cause analysis.
Embodiments of the invention will now be described, by way of example, and with reference to the accompanying drawings, in which:
The drawings and the description of the drawings are of embodiments of the invention and not of the invention itself.
In some of the embodiments, a packet-switched network is monitored via which real-time data is transmitted. Data packets containing real-time data are sniffed by a monitor subagent to monitor a quality-of-service parameter. A monitor agent is notified in response to a threshold breach of the quality-of-service parameter about the breach. Then, a network node manager is notified about the breach, and a root cause analysis is performed.
If a root cause has been detected, a report is sent to a user to inform her/him where the failure originates from. In some of the embodiments, the monitoring of the packet-switched network including a root cause analysis is performed to enable SLA management. Figuring out the root cause of faults occurring in a network may enable an operator to quickly remedy the fault and to enable a service provider to accomplish the fault.
In computer networks and telecommunications, packet switching is now the dominant communications paradigm in which packets (units of information carriage) are individually routed between nodes over data links which might be shared by many other nodes. This contrasts with the other principal paradigm, circuit switching, which sets up a dedicated connection between the two nodes for their exclusive use for the duration of the communication. Packet switching is used to optimize the use of the bandwidth available in a network, to minimize the transmission latency (i.e. the time it takes for data to pass across the network), and to increase the robustness of communication. In packet switching, a file is broken up into smaller groups of data known as packets. Such “packets” carry with them information with regard to their origin, destination and sequence within the original file. This sequence is needed for re-assembly at the file's destination.
In some of the embodiments, the real-time data transmitted over the packet-switched network is data pertaining to a VoIP service. VoIP refers to the routing of voice conversation over the Internet or any other IP-based network. The voice data flows over a general-purpose packet-switched network, instead of traditional dedicated circuit-switched voice transmission lines. VoIP is based on the real-time transport protocol (RTP) which has been designed for transmitting real-time interactive applications, including VoIP and video conferencing. In the ISO-OSI layer model, RTP is part of the application layer, although it is actually a transport protocol which should belong to the OSI-layer 4, the transport layer. Therefore, RTP is also referred to as a transport protocol implemented in the application layer. Via a socket interface, RTP is connected to the UDP (user datagram protocol) which also belongs to the transport layer but is below RTP. Therefore, the sending side encapsulates a media chunk within an RTP packet, then encapsulates the packet in a UDP packet, and then hands the packet to IP layer. The receiving side extracts the RTP packet from the UDP packet, extracts the media chunk from the RTP packet, and then passes the chunk to a media player for decoding and rendering. VoIP makes use of PCM encoding at 64 kbps. Further, it is supposed that the application collects the encoded data in 20 msec chunks, that is 160 bytes in a chunk. The sending side precedes each chunk of the audio data with an RTP header that includes the type of audio encoding, a sequence number, and a timestamp. The RTP header is normally 12 bytes. The audio chunk along with the RTP header form the RTP packet. The RTP packet is then sent into the UDP socket interface. At the receiver side, the application receives the RTP packet from its socket interface. The application extracts the audio chunk from the RTP packet and uses the header fields of the RTP packet to properly decode and play back the audio chunk. It should be mentioned that RTP does not provide any mechanism to ensure timely delivery of data or provide other quality-of-service (QoS) guarantees; it does not even guarantee delivery of packets or prevent out-of-order delivery of packets. RTP allows each source (for example, a camera or microphone) to be assigned its own independent RTP stream of packets. For example, for a video conference between two participants, four RTP streams could be opened—two streams for transmitting the audio (one in each direction) and two streams for transmitting the video (again, one in each direction). However, many popular encoding techniques—including MPEG1 and MPEG2—bundle the audio stream and video stream into a single stream during the encoding process. When the audio stream and video stream are bundled by the encoder, then only one RTP stream is generated in each direction.
RFC 1889 also specifies RTCP (RTP control protocol), a protocol that a networked multimedia application may use in conjunction with RTP. RTP packets are transmitted by each participant in an RTP session to all other participants in the session using IP multicast. For an RTP session, typically there is a single multicast address and all RTP and RTCP packets belonging to the session use the multicast address. RTP and RTCP packets are distinguished from each other by the use of distinct port numbers. (The RTCP port number is set to be equal to the RTP port number plus one.)
RTCP packets do not encapsulate chunks of audio or video. Instead, RTCP packets are sent periodically and contain sender and/or receiver reports that announce statistics that can be useful to the application. These statistics include the number of packets sent, the number of packets lost, and the interarrival jitter. The RTP specification [RFC 3550] does not dictate what the application should do with this feedback information; this is up to the application developer. Senders can use the feedback information, for example, to modify their transmission rates. The feedback information can also be used for diagnostic purposes; for example, receivers can determine whether problems are local, regional, or global.
For each RTP stream that a receiver receives as part of a session, the receiver generates a reception report. The receiver aggregates its reception reports into a single RTCP packet. The packet is then sent into a multicast tree that connects all the session's participants. The reception report includes several fields, the most important of which are listed below.
1. The SSRC (synchronization source identifier) of the RTP stream for which the reception port is being generated.
2. The fraction of packets lost within the RTP stream. Each receiver calculates the number of RTP packets lost divided by the number of RTP packets sent as part of the stream. If a sender receives reception reports indicating that the receivers are receiving only a small fraction of the sender's transmitted packets, it can switch to a lower encoding rate, with the aim of decreasing network congestion and improving the reception rate.
3. The last sequence number received in the stream of RTP packets.
4. The interarrival jitter, which is a smoothed estimate of the variation in interarrival time between successive packets in the RTP stream.
For each RTP stream that a sender is transmitting, the sender creates and transmits RTCP sender report packets. These packets include information about the RTP stream, including:
The User Datagram protocol (UDP), on which RTP is based, is one of the core protocols of the Internet protocol suite. Using UDP, programs on networked computers can send short messages known as datagrams to one another. UDP does not provide the reliability and ordering guarantees that TCP (Transmission Control Protocol) does; datagrams may arrive out of order or go missing without notice. However, as a result, UDP is faster and more efficient for many lightweight or time-sensitive purposes. Also its stateless nature is useful for servers that answer small queries from huge numbers of clients.
UDP is a minimal message-oriented transport layer protocol that is documented in RFC 768. In the TCP/IP model, UDP provides a very simple interface between a network layer below and an application layer above. UDP provides no guarantees for message delivery and a UDP sender retains no state on UDP messages once sent onto the network. Lacking reliability, UDP applications must generally be willing to accept some loss, errors or duplication. Some applications such as TFTP may add reliability mechanisms into the application layer as needed. Most often, UDP applications do not require reliability mechanisms and may even be hindered by them. If an application requires a high degree of reliability, a protocol such as the Transmission Control Protocol (TCP) or erasure codes may be used instead.
The term “sniffing” refers to a software or hardware entity that receives the data traffic of a network, records the data traffic and evaluates the data traffic afterwards. In other words, a sniffer is a tool for analyzing a local area network. Sniffing may be performed in promiscuous mode and non-promiscuous mode. In the non-promiscuous mode, only the data traffic of its own computer is controlled, whereas in promiscuous mode, the sniffer also “sees” the data traffic which is determined for another system. Furthermore, the data that can be seen by a sniffer depends on the network structure. If computers are used with hubs, then all the traffic can be seen by the other sniffers, since all computers belong to the same collision domain. If, however, a switch is used, then only little or no data traffic might be seen by the sniffer which is not determined for the system to be sniffed. All of the traffic in the designated Ethernet broadcast domain can be seen if the sniffer sniffs a mirror port that replicates all traffic seen across the ports of the switch in question. Most switch vendors provide support for mirror ports. However, configuring a mirror port is a performance penalty on the switch, unless there is dedicated hardware to accomplish mirroring. One example of a sniffer is “Ethereal” which uses Winpcap utilities to sniff and capture UDP packets and parses them for encapsulated RTP/RTCP packets.
The term “network node manager (NNM)” refers to a program or hardware entity that is able to monitor network nodes of a network. An example of a network node manager is Hewlett Packard's OpenView Network Node Manager (OV-NNM) having components, such as snmpCollect, ovwdb, topodb, an event subsystem, trap-to-event conversion and a path analysis subsystem, as will be explained below.
In some of the embodiments, the real-time data packets being sniffed are control packets that accompany the RTP payload. These control packets are the above-mentioned RTCP data packets which contain control information about the RTP data traffic.
In some of the embodiments, the quality-of-service parameter refers to end-to-end egress delay and ingress delay which give a quantitative measure of total latency in signal transmission which is propagation plus insertion plus other delays in each direction (one way delay). A delay value above a certain threshold will cause inability to comprehend peer user's response during a VoIP-call.
In other embodiments, the quality-of-service parameter refers to echo perceived by the VoIP user (roundtrip delay). A delay value above a certain threshold may cause the user to playback his own transmitted media. This delay manifests itself on a bad telephone line by the speaker's own voice being heard again (much like an echo).
In some of the embodiments, the quality-of-service parameter refers to jitter. A crucial component of end-to-end delay is the random queuing delays in the routers of a network. Because of these varying delays within the network, the time that elapses between a packet being generated at the source and its arrival at the receiver can fluctuate from packet to packet. This phenomenon is called jitter. As an example, let us consider two consecutive packets within a talk spurt in a VoIP application. The sender sends the second packet 20 msecs after sending the first packet. But at the receiver, the spacing between these packets may become greater than 20 msecs. To see this, suppose the first packet arrives at a nearly empty queue at a router, but just before the second packet arrives at the queue a large number of packets from other sources arrive at the same queue. Because the first packet suffers a small queuing delay and the second packet suffers a large queuing delay at this router, the first and second packets become spaced by more than 20 msecs. The spacing between consecutive packets can also become less than 20 msecs. To see this, again we can consider two consecutive packets within a talk spurt. It is assumed that the first packet joins the end of a queue with a large number of packets, and the second packet arrives at the queue before packets from other sources arrive at the queue. In this case, the two packets find themselves one right after the other in the queue. If the time it takes to transmit a packet on the router's outbound link is less than 20 msecs, then the first and second packets become spaced apart by less than 20 msecs. If the receiver ignores the presence of jitter and plays out chunks as soon as they arrive, then the resulting audio quality can easily become unintelligible at the receiver. Jitter may often be removed by using sequence numbers, timestamps, and a playout delay, as discussed below.
In some of the embodiments, the quality-of-service parameter refers to “throughput”, which determines whether the network nodes are capable of meeting the high bandwidth requirement of various RTP transmissions. If throughput is exceeded, this usually translates into lost packets and is manifested as signal loss to the VoIP user. Conversely, a low throughput would indicate under-utilization or over-capacity of network nodes.
In other embodiments, the quality-of-service parameter refers to “instantaneous signal loss” which is perceived as sudden spurts of signal loss by the user and a corresponding signal-to-noise ratio. This phenomenon manifests itself when syllables in our speech are dropped, while using a non circuit-switched or wireless telephone line.
In some of the embodiments, the quality-of-service parameter refers to “accumulated content loss” which has billing and accounting implications. Users are provisioned for a certain throughput at a certain demanded bandwidth. Users are usually billed for the net content transferred or the bandwidth used. These computations will be governed by this metric.
In other embodiments, notifying a network node manager comprises sending a trap to the network node manager. In some of the embodiments, this trap is an SNMP trap which is generated by using the command “snmpnotify”. The trap conveys information concerning the threshold breach to the network node manager where a root cause analysis is performed.
In some of the embodiments, performing a root cause analysis includes performing a path analysis to find out the path over which the real-time data is transmitted.
In other embodiments, the root cause analysis comprises examining interfaces of network nodes lying on the path found out. The term “interface” as used herein refers to the connections of a network node to other network nodes or parts of a network (subnet). A router, for example, has interfaces to all subnets to which it is connected.
In some of the embodiments, the root cause analysis is performed in cooperation with the monitor subagent which detected the threshold breach.
In other embodiments, an alarm is triggered in response to breaching a threshold. If a breach of a quality-of-service parameter is forwarded to a media gateway controller which triggers media gateways to renegotiate bandwidth of the communication, which may result in providing a higher bandwidth of a VoIP-call.
In some of the embodiments, sniffing is initiated by means of a media gateway controller, more particularly, the sniffing is initiated by signaling messages issued by the media gateway controller.
In other embodiments, the network node manager is coupled to a data warehouse which stores data concerning threshold breaches over a longer period of time and therefore enables a user to make statistical analysis concerning threshold breaches. A data warehouse may also provide an appropriate database for enabling a user to perform “what-if” analysis. In a “what-if” analysis, a user may provide network nodes and their performance data and determine the probability of a threshold breach under these specific circumstances.
Returning now to
In the example network shown in
In the example, a VoIP-call is set-up between an analog phone 8.2 behind a residential gateway 5 and a public switched telephone network user connected on a trunk interface on the trunking gateway 4. Network performance is monitored for transport of RTP/RTCP packets in the connections which are indicated as dotted lines and which are also referred to as media paths. Monitoring beyond the trunking gateway 4, i.e. in the public switched telephone network 6, is irrelevant since it is a circuit switched domain. Measurement beyond the residential gateway 5 along paths towards phone 8.2 is also irrelevant because it is an analog media connection. A path analysis in the presence of a network node manager's discovered topology data will yield the RTP/RTCP propagation paths. Throughput at each interface could give a measure of throughput of the VoIP media traffic in the presence of other traffic. A percentage utilization of these interfaces can also be arrived at. A user can visualize the network path used for the call set up and be able to pinpoint which interfaces cause the packet loss. If an end-to-end packet loss is reported, it can be correlated to the interfaces causing packet loss by looking up the analyzed path and consulting a network node manager topology database to resolve the associated interfaces. Thus, a root cause analysis is possible. If jitter or round trip delay threshold violations are detected, a user can be alerted via a network node manager event subsystem about the degradations. Also, these events can be forwarded to a subscribing media gateway controller 3, which uses the feedback to adapt to the changed network conditions and direct the media entities (in this case residential gateway 5 and trunking gateway 4 peers) to e.g. renegotiate a lower bandwidth communication. A user can profile the performance metric data collected to a data warehouse for trend analysis. A “what-if” analysis can be applied to that data by varying metric derived parameters to see how other metrics are affected. This is useful in capacity planning. For example, an increasing trend in throughput across certain interfaces at certain times of the day suggests that more capacity needs to be provisioned at that time of the day. Roundtrip delay values could indicate the efficacy of paths taken by the UDP packets carrying the RTP payload data. A user can devise SLA events based on the collected metrics and define corresponding accounting events to be forwarded to a billing system.
The analog phones 41 are connected to the network via a media gateway 42 which acts as a translation unit between disparate telecommunication networks such as PSTN (public switched telephone network) and the IP-based Internet. Media gateways 42 enable multimedia communications across next generation networks over multiple transport protocols such as ATM and IP. The media gateway 42 is controlled by a softswitch 43 which provides the call control and signaling functionality.
It should be mentioned that the terms “softswitch” and “media gateway controller” are similarly used; however, there is a difference in functionality. If a VoIP master/slave signaling protocol is in use for both endpoints, such as MGCP (media gateway control protocol), in that case the MGC will have to drive the signaling for phones sitting behind a media gateway, e.g. a residential gateway or an access gateway in
Communication between media gateways and softswitches is achieved by means of protocols such as the MGCP mentioned above. Media gateways 42.1 and 42.2 perform the conversion from TDM (time division multiplexing) voice to VoIP.
The network shown in
The network shown in
Furthermore, it should be noted that two IP phones 48.1 and 48.2 are provided in the network. These IP phones need not be connected to the network via the media gateways 42.1 and 42.2 since they are enabled to send and receive IP packets directly.
After the telephone call 53 has been set up, monitor subagent 44.2 is triggered by the softswitch 43 to check the data traffic on its broadcast domain. The monitor subagent 44.2 may be considered as a plug-in module that is resident in an intermediate node in the network path of the RTP packets. This node is able to capture UDP packets carrying RTP payload and pass them upwards to the plug-in module. The monitor subagent 44.2 is a node that is able to sniff Ethernet frames promiscuously in an appropriate Ethernet broadcast domain. The monitor subagent 44.2 is configured with threshold values for various quality-of-service parameters that are collected and measured. The monitor subagent 44.2 is able to communicate with the network node manager 47 to send notifications if thresholds are breached. The monitor subagent 44.2 accepts commands, by the softswitch 43, to start and stop sniffing the network traffic so that there is no unnecessary sniffing in the absence of RTP traffic. The monitor subagent 44.2 may follow signaling messages issued by the softswitch 43 and infer when and which peers to monitor for RTP traffic. In the example, the monitor subagent 44.2 detects that incoming packets are less than outgoing packets. This ingress packet loss may be determined by means of the RTCP packets sniffed by the monitor subagent 44.2 in the broadcast domain. The RTCP packet contains a reception report which indicates the number of RTP packets lost divided by the number of RTP packets sent as part of the stream. The monitor subagent 44.2 compares the ingress packet loss with a threshold which has been defined by a user. The threshold breach is then reported to the monitor agent 46, which in turn sends an SNMP-trap to the network node manager (NNM) 47. A trap is an unsolicited report about an event which is sent from an SNMP agent 54 resident on the monitor agent 46 to the SNMP-manager 53 resident on the network node manager 47. If the SNMP-manager 53 on the network node manager 47 receives a trap, the path analysis subsystem 49 on the network node manager 47 triggers a root cause analysis.
The timestamp field is 32 bits long. It reflects the sampling instant of the first byte in the RTP data packet. A receiver can use timestamps in order to remove packet jitter introduced in the network and to provide synchronous playout at the receiver. The timestamp is derived from a sampling clock at the sender. As an example, for audio the timestamp clock increments by one for each sampling period. A 32-bit synchronization source identifier field 32 is provided which identifies the source of an RTP stream. Typically, each stream in an RTP session has a distinct SSRC. The SSRC is not the IP address of the sender, but instead is a number that the source assigns randomly when the new stream is started. The probability that two streams get assigned the same SSRC is very small. Should this happen, the two sources pick a new SSRC value.
On the basis of the trace route analysis, the root cause analysis subsystem 51 determines faulty interfaces which are responsible for the packet loss. To this end, the interfaces on the path are identified and it is asked via the command snmpCollect for the variables “InterfaceInOctets”, “InterfaceOutOctets”, “InterfaceErrorOctets” stored in the MIBs of the individual network components of the actual path.
Once the event translated from INGRESS_PACKETLOSS_EXCEEDED trap is received by the path analysis subsystem 49, it does the following things:
At a) it looks up NNM topology database 52 to get a handle of the node object corresponding to the source and destination IP addresses. Once the node objects are determined, using their ID (Id of receiver/sender), a dump of the network path (layer-2/layer-3) between the source and the destination is obtained. This yields all the participating network nodes (switches, routers, etc.) in that path that carry the traffic between the endpoints. Now the path analysis subsystem 49 walks through each network node and retrieves the associated interfaces that are actually carrying the traffic. An operational/administrative status of the interfaces is also available at this time based on status last known by NNM 47.
At b), if one or more interfaces are found to be operationally or administratively “down”, the root cause is determined as status ‘down’, interfaces [ifID][ifID] . . . ” and then an enriched event is generated based upon the root cause and propagated to the event subsystem 50 for correlation and reporting. If this is not the case, an NNM snmpCollect component is consulted to assess the traffic performance of the interfaces concerned. If it is known that none of the interfaces are “down” but packet loss is still experienced nevertheless, it may be that packets are dropped at the interfaces due to congestion or error. Or it could be that the network path has changed dynamically to a more “lossy” or “down” path since network node manager 47 last knew the path between the endpoints. Interface performance data returned by snmpCollect will help assess whether the interfaces are performing poorly leading to the loss or not. It may be noted that snmpCollect collects the performance data from nodes via SNMP in an asynchronous manner, to be ready with the data when requested by path analysis. Therefore it is essential that snmpCollect is able to maintain stable SNMP communication with the “Management IP address” of the nodes inspite of the nodes' individual interfaces being disruptive. If it is assessed that interfaces are faulty, the root cause is determined as “faulty interface(s) [ifID][ifID] . . . ” and an enriched event is generated based upon the root cause and propagated to the event subsystem 50 for correlation and reporting.
At c), the path analysis subsystem 49 then instructs the observing monitor subagent 44.2 to attempt an ICMP-based trace-route of the affected path. The trace-route is done by the sub-agent because it is able to see the same layer-3 path as seen by the endpoint experiencing the loss. This may or may not be successful because the path may already be sufficiently lossy or down not to be able to accommodate additional trace-route traffic. If trace-route is successful, path analysis is able to compare the path with the path last known by the network node manager 47. If a new path is indicated, path analysis again resolves the nodes in the path, their interfaces, their status and if necessary consults snmpCollect to assess their health. If faulty interfaces are identified in the new path, root cause is determined as “path changed, faulty interface(s) [ifID][ifID] . . . in new path”, path analysis sends an enriched event to the event subsystem with root cause analysis identifying faulty interface(s) causing packet loss.
At d), else the path analysis subsystem 49 now revisits the switches and routers in the last known path between the endpoints. Using SNMP on only the management IP addresses of the devices, it attempts to determine if the path has changed by analyzing neighbors reported by the devices. For example, with regard to a router, if the “next hop” IP address to route the destination has changed this is an indication the layer-3 path has changed. For a switch, a combined analysis of its 802.1d forwarding table and any neighbor advertisement table such as CISCO CDP or EXTREME EDP may yield any layer-2 path that has changed. If the path is changed, a new discovery and complete fault monitoring is scheduled for NNM, and the root cause for now is determined as “path changed, unknown interface behavior in new path”.
At e), finally, an enriched event is generated on the basis of the root cause and propagated to the event subsystem for correlation and reporting.
The “enriched event” contains the result of a root cause analysis based on input parameters carried by the original input event(s) and further analysis done by various subsystems acting on the input parameters. This event may also have some parameters canonicalized within the network node manager domain. In the example of ingress packet loss provided, after the path-analysis subsystem consumes the incoming event, it calls on the help of various other subsystems as follows—network node manager ovwdb/topodb to identify the network interfaces concerned, snmpCollect to determine interface health, comparing trace-route data as reported by the subagent against the last stored topology to arrive at the root cause. In the process of resolving affected interfaces, it also canonicalizes the endpoints by replacing Source and Destination IP Addresses in the incoming event by node object IDs held in the network node management topology database 52.
Thus, the embodiments of the invention described above allow for an sniffing-based network monitoring.
All publications and existing systems mentioned in this specification are herein incorporated by reference.
Although certain methods and products constructed in accordance with the teachings of the invention have been described herein, the scope of the coverage of this patent is not limited thereto. On the contrary, this patent covers all embodiments of the teachings of the invention fairly falling within the scope of the appended claims either literally or under the doctrine of equivalents.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US6766325 *||Dec 2, 1999||Jul 20, 2004||Microsoft Corporation||System and method for maintaining data for performing “what if” analysis|
|US7580994 *||Sep 10, 2004||Aug 25, 2009||Nortel Networks Limited||Method and apparatus for enabling dynamic self-healing of multi-media services|
|US20020051464 *||Aug 14, 2001||May 2, 2002||Sin Tam Wee||Quality of transmission across packet-based networks|
|US20030128692 *||Jan 4, 2002||Jul 10, 2003||Mitsumori Derek Hisami||Voice over internet protocol (VoIP) network performance monitor|
|US20030223381 *||Jun 4, 2002||Dec 4, 2003||Osmo Schroderus||Method for controlling parties in real-time data communication|
|US20040024865 *||Jul 30, 2002||Feb 5, 2004||Jiandong Huang||Method and apparatus for outage measurement|
|US20040114570 *||Nov 22, 2003||Jun 17, 2004||Vikberg Jari Tapio||Combining narrowband applications with broadband transport|
|US20040181709 *||Mar 13, 2003||Sep 16, 2004||Sun Microsystems, Inc.||Method, system, and article of manufacture for fault determination|
|US20040264433 *||Oct 24, 2002||Dec 30, 2004||Diego Melpignano||Wireless communication arrangements with header compression|
|US20050068942 *||Sep 30, 2003||Mar 31, 2005||Chu Thomas P.||Method and apparatus for establishment and management of voice-over IP virtual private networks in IP-based communication systems|
|US20050094651 *||Sep 29, 2004||May 5, 2005||Alcatel||Lawful interception gateway|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8284688 *||Jun 29, 2009||Oct 9, 2012||Fujitsu Limited||Packet capture apparatus, packet capture method, and computer readable medium having a packet capture program|
|US8300630 *||Oct 30, 2012||International Business Machines Corporation||UPD-based soft phone state monitoring for CTI applications|
|US8402311 *||Mar 19, 2013||Microsoft Corporation||Monitoring activity with respect to a distributed application|
|US8437266 *||May 7, 2013||Avaya Inc.||Flow through call control|
|US8631154 *||Jun 29, 2011||Jan 14, 2014||International Business Machines Corporation||Dynamically modifying quality of service levels for resources in a networked computing environment|
|US8661295 *||Mar 31, 2011||Feb 25, 2014||Amazon Technologies, Inc.||Monitoring and detecting causes of failures of network paths|
|US8937870||Sep 11, 2012||Jan 20, 2015||Amazon Technologies, Inc.||Network link monitoring and testing|
|US8971193 *||Oct 24, 2011||Mar 3, 2015||Edgewater Networks, Inc.||Method to determine media paths in a SIP network using information from endpoints and intermediate devices|
|US9001667||Mar 31, 2011||Apr 7, 2015||Amazon Technologies, Inc.||Monitoring and detecting causes of failures of network paths|
|US9025462||Jan 14, 2013||May 5, 2015||Qualcomm Incorporated||Reception report aggregation|
|US9065772||Jan 13, 2014||Jun 23, 2015||International Business Machines Corporation||Dynamically modifying quality of service levels for resources running in a networked computing environment|
|US9104543||Apr 6, 2012||Aug 11, 2015||Amazon Technologies, Inc.||Determining locations of network failures|
|US9197495||Feb 11, 2013||Nov 24, 2015||Amazon Technologies, Inc.||Determining locations of network failures|
|US9210038||Feb 11, 2013||Dec 8, 2015||Amazon Technologies, Inc.||Determining locations of network failures|
|US9246747||Nov 15, 2012||Jan 26, 2016||Hong Kong Applied Science and Technology Research Co., Ltd.||Adaptive unified performance management (AUPM) with root cause and/or severity analysis for broadband wireless access networks|
|US9313107||Apr 28, 2015||Apr 12, 2016||International Business Machines Corporation||Dynamically modifying quality of service levels for resources running in a networked computing environment|
|US9385917||Mar 31, 2011||Jul 5, 2016||Amazon Technologies, Inc.||Monitoring and detecting causes of failures of network paths|
|US20050199292 *||Mar 10, 2005||Sep 15, 2005||Stedman David W.||Fluid device actuator with manual override|
|US20090232127 *||Mar 14, 2008||Sep 17, 2009||Peeyush Jaiswal||UPD-Based Soft Phone State Monitoring for CTI Applications|
|US20100020708 *||Jun 29, 2009||Jan 28, 2010||Fujitsu Limited||Packet capture apparatus, packet capture method, and computer readable medium having a packet capture program|
|US20110051605 *||Aug 26, 2009||Mar 3, 2011||Avaya Inc.||Flow through call control|
|US20120017120 *||Jul 19, 2010||Jan 19, 2012||Microsoft Corporation||Monitoring activity with respect to a distributed application|
|US20130007234 *||Jan 3, 2013||International Business Machines Corporation||Dynamically modifying quality of service levels for resources in a networked computing environment|
|US20130100859 *||Oct 24, 2011||Apr 25, 2013||David Samuel Martin||Method to determine media paths in a SIP network using information from endpoints and intermediate devices|
|WO2013007180A1 *||Jul 9, 2012||Jan 17, 2013||Huawei Technologies Co., Ltd.||Method, device and system for network congestion control|
|Cooperative Classification||H04L65/104, H04L65/1036, H04L65/103, H04L65/1026, H04L41/0213, H04L41/0681, H04L41/5003, H04L65/80, H04L12/2602, H04L43/0829, H04L41/5038, H04L43/087, H04L43/12, H04L41/0226, H04L43/00, H04L43/06, H04L29/06027, H04L43/16, H04L43/045, H04L43/14|
|European Classification||H04L29/06M8, H04L29/06C2, H04L29/06M2N2M4, H04L29/06M2N2M2, H04L29/06M2N2S4, H04L29/06M2N2S2, H04L12/26M, H04L65/80, H04L43/00, H04L41/50A, H04L41/50F|
|Mar 30, 2007||AS||Assignment|
Owner name: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P., TEXAS
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:DAS CHOUDHURY, JONMEJOY;REEL/FRAME:019176/0813
Effective date: 20070318
|Sep 13, 2011||CC||Certificate of correction|
|Oct 23, 2014||FPAY||Fee payment|
Year of fee payment: 4
|Nov 9, 2015||AS||Assignment|
Owner name: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP, TEXAS
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.;REEL/FRAME:037079/0001
Effective date: 20151027